What mechanisms does Kumo provide to detect data leakage?

Kumo provides severals tools and mechanisms to detect data leakage. Column analysis in particular provides an effective mechanism for detecting data leakage, since columns with variation of prediction values dramatically higher than any others are tell-tale signs of data leakage. For example, if you are predicting whether a customer will churn in the next 30 days and you have a column indicating whether the customer will cancel their subscription in the next 30 days, this would constitute both target leakage and future information leakage; in this case, you would look for an excessively high contribution score for this column.

More Information:

• How do I handle time correctness to prevent data leakage?
• How can I improve the quality of my data?
• Explainability
• Column Analysis